Present and Future of AI in Renewable Energy

Domain : A Comprehensive Survey

Abdur Rashid1 , Parag Biswas2 , Angona Biswas3 ,
MD Abdullah Al Nasim4*, Kishor Datta Gupta5 , Roy George6
1, 2 MSEM Department, Westcliff university, California, United States.
3, 4 Researchand Development Department, Pioneer Alpha, Dhaka,
Bangladesh.
arXiv:2406.16965v1 [cs.LG] 22 Jun 2024

5, 6 Department of Computer and Information Science, Clark Atlanta

University, Georgia, USA.

*Corresponding author(s). E-mail(s): [email protected];

Contributing authors: [email protected];
[email protected]; [email protected]; [email protected];
[email protected];

Abstract
Artificial intelligence (AI) has become a crucial instrument for streamlining pro-
cesses in various industries, including electrical power systems, as a result of
recent digitalization. Algorithms for artificial intelligence are data-driven models
that are based on statistical learning theory and are used as a tool to take use
of the data that the power system and its users generate. Initially, we perform a
thorough literature analysis of artificial intelligence (AI) applications related to
renewable energy (RE). Next, we present a thorough analysis of renewable energy
factories and assess their suitability, along with a list of the most widely used and
appropriate AI algorithms. Nine AI-based strategies are identified here to assist
Renewable Energy (RE) in contemporary power systems. This survey paper com-
prises an extensive review of the several AI techniques used for renewable energy
as well as a methodical analysis of the literature for the study of various intelli-
gent system application domains across different disciplines of renewable energy.
This literature review identifies the performance and outcomes of nine different
research methods by assessing them, and it aims to distill valuable insights into
their strengths and limitations. This study also addressed three main topics: using
AI technology for renewable power generation, utilizing AI for renewable energy
forecasting, and optimizing energy systems. Additionally, it explored AI’s supe-
riority over conventional models in controllability, data handling, cyberattack

1
 prevention, smart grid implementation, robotics- AI’s significance in shaping the
future of the energy industry. Furthermore, this article outlines future directions
in the integration of AI for renewable energy.

Keywords: Artificial Intelligence (AI), renewable energy, Machine Learning, Deep

Learning, Power Consumption, Power System

1 Introduction
The fundamental component of financially sound, ecologically responsible, and sus-
tainable electricity generation is renewable energy, or RE. The role that renewable
energy plays in reducing climate change [1], [2], [3] is among the most compelling argu-
ments for adopting it. When generating energy, renewable energy sources like solar,
wind, and hydroelectric power emit little to no greenhouse gases, in contrast to fossil
fuels [4]. When it comes to security and sustainability, renewable energy sources are
superior to limited fossil fuels [5]. Natural replenishment of renewable resources, such
as sunshine, wind, and water, makes them a steady and dependable energy source free
from the price volatility and geopolitical unpredictability of fossil fuels [6]. By sup-
plying power to isolated and neglected areas, renewable energy technologies provide a
viable way to combat energy poverty [7]. Off-grid solar systems may provide rural com-
munities with dependable and reasonably priced power, raising living standards and
promoting economic growth. Not only is renewable energy crucial, but it is also neces-
sary for a thriving and sustainable future. We can solve a number of issues, including
as public health, economic growth, energy security, and climate change, while build-
ing a more resilient and sustainable environment for coming generations by embracing
renewable energy sources and moving toward a clean energy economy [8], [9].
According to an official report by the International Energy Agency (IEA), the
need for fossil fuels to generate power has begun to decline since 2019 [10], coincid-
ing with the rise in the use of renewable energy sources to meet the world’s energy
needs. In an effort to improve the efficiency of energy conversion, particularly in terms
of RE generation, research on RE technologies is constantly expanding. The current
global demand, particularly in developed and developing nations, necessitates the
replacement of traditional electrical generation resources like fossil fuels with more
sustainable energy technology [11]. Global warming and change in the climate are two
harmful environmental problems brought on by fossil fuel-based energy sources. Over
the past few decades, there has been an exponential increase in the greenhouse gas
emissions from power generation into the environment. In order to tackle the current
environmental issue, renewable energy (RE) technologies including solar, wind, hydro,
biomass, geothermal, and hydrogen energies have been adopted to create electricity
[12]. It is anticipated that in the future, renewable energy sources will provide all of
the electricity needed to power residential areas. Data regarding Japan’s 2018 renew-
able energy percentage is available from the Institute for Sustainable Energy Policies
(ISEP) is shown in Figure 1.

2
 International Renewable Energy Agency (IRENA) and Renewable Energy Policy
Network for the 21st Century (REN21) [13] provides the important information about
Renewable Energy which is recorded in Table 1.

Table 1 Global Renewable Energy Capacity by Source (2020-2023)

Year Hydro (GW) Wind (GW) Solar (GW) Bioenergy (GW) Geothermal (GW)
2020 1305 732 707 121 14
2021 1300 780 900 130 13
2022 1350 850 1000 140 15
2023 1400 900 1100 150 16

Fig. 1 The data in the pie chart shows the amount of energy that Japan’s power transmission and
distribution networks transported in 2018. [14]
The global pandemic of the corona virus made it difficult to obtain current data
on the world’s electrical generation in 2021 from a number of sources, as [10]’s search
revealed. Figure 2 illustrates how the difference in energy generation from fossil fuels
and renewable energy sources has shrunk between 2010 and 2020. From 121,531 TWh
in 2010 to 136,131 TWh in 2019, the amount of fossil fuels used to generate electricity
climbed somewhat, but it then declined in 2020. Conversely, the amount of energy
generated by renewable sources has grown dramatically, rising from 4098 TWh in 2010
to 7,140 TWh in 2019. In contrast to the production of fossil fuels, renewable energy
(RE) accounted for only 3.26–5.60

3
Fig. 2 A comparison between the growth of renewable energy sources and fossil fuel generation
(2010–2020) [10]
Figure 3 is showing forcasting that in 2025 the production of Renewable Energy
will be reached to 521.95 TWh in USA. This value is incredible with compared to
China, Japan, Brazil and Russia’s RE production [15].

Fig. 3 TWh of data on the generation of renewable energy in the United States from 2016 to 2019
and from 2020 to 2025. Authors’ calculations serve as the source. [16]

1.1 Relevance of Renewable Energy

Renewable energy is extremely important in many aspects that are vital to sustainable
development [17]. First and foremost, it plays a critical role in reducing greenhouse
gas emissions, since renewable energy sources such as hydroelectric, solar, and wind
power emit less greenhouse gases and provide a sustainable substitute for fossil fuels
[18].In terms of the economy, the renewable energy sector supports local economies
and promotes the shift to a green economy by generating employment, advancing
technological innovation, and drawing investment [17]. Furthermore, the utilization of
renewable energy improves public health by mitigating air and water pollution that

4
arises from burning fossil fuels, consequently relieving respiratory ailments and other
health-related issues. Most importantly, it powers socioeconomic growth by enabling
fair access to electricity, especially in isolated and underprivileged areas. In addition,
the shift to renewable energy drives technological improvement by promoting energy
storage, grid integration, and efficiency research and development [19].

Fig. 4 Sustainable eco-friendly solutions for power [20]

When discussing the issues posed by climate change, many of us immediately
consider the impending energy crisis that we will almost certainly confront. Fossil fuel
resources are getting harder to come by, but global CO2 emissions are still rising.
Future energy requirements and security need the broad use of new renewable energy
sources (shown in Figure 4). That still calls for a radical transformation in the way
that scientists, students, and world leaders see renewable energy and climate change.
Since renewable energy is the way of the future, we must refocus on sustainability via
innovation by learning about energy, chemistry, and technology.

1.2 Renewable Energy Sector without the influence of

Artificial Intelligence
Before the integration of artificial intelligence (AI) into the energy sector, renew-
able energy technologies such as solar, wind, hydroelectric, geothermal, and biomass
were already established and evolving [21]. Developments in these technologies pri-
marily focused on improving efficiency, reducing costs, and addressing environmental
concerns. However, despite the advancements brought by AI, there were some lim-
itations and challenges. One of the main lacking aspects was the dependency on
traditional methods for data analysis and decision-making, which often lacked real-
time insights and predictive capabilities [21]. Additionally, the integration of AI into
renewable energy systems required significant investment in infrastructure, data col-
lection, and training algorithms, which posed barriers to adoption, particularly in less
developed regions. Moreover, there were concerns regarding data privacy, security, and
the potential for AI algorithms to exacerbate biases or inequalities in energy access and

5
distribution. Despite these challenges, the integration of AI has the potential to revo-
lutionize the renewable energy sector by optimizing operations, improving efficiency,
and accelerating the transition to a sustainable energy future. At that time researcher
used to focus on improvement of renewable energy sources by means of developing
tools, energy management system, finding proper solution for those which are harm-
ful for environment. [22] this research article focuses on assessing the present status
of electricity access, identifying issues and challenges, and evaluating the effective-
ness of renewable energy and energy efficiency in improving electricity access in urban
poor areas, specifically in the Korail slum of Dhaka (2017). This research outcome
suggests that rooftop solar PV, electricity generation from solid waste, and replac-
ing incandescent lamps with CFLs are promising approaches to enhance electricity
access in the Korail slum, with significant potential for surplus electricity generation
and cost savings. A separate study [23] examines the impact of environmental regula-
tions on sustainability, taking into account factors such as research and development
in renewable energy (RERD), advancements in technology (TI), and the overall eco-
nomic performance (GDP). The researchers employed wavelet methods, a specialized
mathematical tool, to examine the interconnections between these parameters over
various time periods and in diverse manners. They discovered that increasing funding
for RERD, implementing more stringent environmental regulations, and promoting
the adoption of new eco-friendly technology are crucial for safeguarding the environ-
ment. Additionally, they discovered significant correlations between CO2 emissions
and factors such as technological advancements and economic expansion.

1.3 Artificial Intelligence in the Renewable Energy Sector

Innovations in digital technology have the power to fundamentally alter our energy
supply, commerce, and consumption. The artificial intelligence (AI) technology powers
the new digitalization paradigm [24]. Intelligent software that maximizes operations
and decision-making will autonomously manage the integration of energy supply,
demand, and renewable sources into the power grid. AI will be essential to reach-
ing this objective. A lot of data may be gathered by connecting the electricity grid
to various devices and sensors. When combined with AI, this data can provide grid
operators with fresh perspectives for more effective operation control. It gives energy
providers the freedom to deftly modify supply in response to demand. AI can support
distributed energy management and microgrid integration [25]. It is challenging to
maintain a balance in the energy flow inside the grid when community-scale renewable
energy producing units are integrated with the main grid.The quality and conges-
tion problems may be greatly helped by the AI-powered control system. Intelligent
Energy Storage (IES) and artificial intelligence combined have the potential to offer
the renewable energy sector a dependable and long-lasting solution [25].
The effects of AI on the business and energy sectors are depicted in Figure 5.
The influence of AI on energy enterprises across several industries is expected to
exceed current expectations [24]. The influence of AI technology on various business
kinds over the next five years is covered by the portion of the red line in Figure
5. The vertical axis illustrates the ”Effect of processes,” while the horizontal axis
explains the ”Effect of offerings.” The ”Effect of Offerings” offers more chances to

6
experience the influence of AI in many industries (accept or reject as desired), while
the ”Effect of Processes” refers to a sequence of events or procedures followed in order
to accomplish a certain objective. The majority of firms anticipate greater effects on
supply chain management, production and operations, customer-focused activities,
and energy information technology (IT).

Fig. 5 Effects of AI in the business along with energy sectors [24]

Control solutions that make use of data-driven models of the wind-photovoltaic
system employ AI and ML algorithms (Shown in Figure 6). A wind turbine’s power
production may be predicted by an AI-based control algorithm based on meteorological
factors like wind direction and speed [26]. Similar to this, a control system based on
machine learning may be created to forecast a photovoltaic panel’s power production
depending on variables like temperature and sun irradiation. The wind-PV system’s
performance may then be optimized using these forecasts.

7
Fig. 6 AI and ML-based PV-wind system control [26]

Data-driven models such as support vector machines (SVMs), k-nearest neighbors

(k-NNs), regression trees, multiple linear regression (LR), gradient boosting (GB),
and random forest (RF) are commonly used in the field of renewable energy (RE)
prediction, leveraging machine learning (ML) techniques. Deep learning models such
as long short-term memory (LSTM), gated recurrent units (GRUs), and deep neural
networks (DNNs) have been used to predict power consumption and renewable energy
(RE) generation across different time periods, yielding outstanding results [27], [28].
In addition, the combination of LSTM with autoencoders (AE) has produced positive
outcomes [29].

1.4 Survey Framework

This research focuses on the application of AI methods in the energy industry. In order
to enable researchers and readers to evaluate each other’s AI endeavors, goals, fresh,
cutting-edge applications, difficulties, and future paths, this study attempts to provide
a realistic baseline. Three main topics were covered: (i) utilising current developments
in AI technology to create renewable power; (ii) use AI to forecast renewable energy;
and (iii) optimizing energy systems. Furthermore, controllability, large data handling,
cyberattack prevention, smart grid, IoT, robotics, energy efficiency optimization, pre-
dictive maintenance control, and computational efficiency were all areas in which this
study examined how AI approaches surpass conventional models. The importance of
AI in the next energy industry is another major topic of this study.

8
Fig. 7 Survey framework of proposed survey research.
The survey structure of this research is organized in a strategic and logical way to
cover all the important issues which is shown in the Figure 7. The rest of the paper is
organized in this way.

2 AI Driven Methods For Renewable Energy

2.1 Forecasting Renewable Energy Generation using Deep
Learning
Authors [30] proposes the ’AB-Net’ architecture, combining autoencoder (AE) and
bidirectional long short-term memory (BiLSTM) for one-step forecasting of renewable
energy (RE) generation. By acquiring data from various RE sources and deep prepro-
cessing, the model extracts features and forecasts RE generation. Two datasets that
are publicly accessible were employed in order to test and assess the efficacy of the
suggested method: the wind dataset and the solar dataset. The model consists of three
main steps: data acquisition from various renewable energy sources, deep preprocess-
ing to clean and normalize the acquired data, and feature extraction and forecasting
using AE and BiLSTM which is shown in Figure 8.

ht = σh (Wxh Xt + Whh ht−1 + bh ) (1)

yt = σy (Why ht + by ) (2)
The state computation of a hidden layer in a Long Short-Term Memory (LSTM)
network is explained in Equation (1). At each time step t, the hidden state ht is
updated and maintained based on the previous hidden state ht−1 and the current input
xt to the layer. In Equation (1), σh denotes the activation function, Wxh is the weight
matrix from the input to the hidden layer, Whh is the weight matrix for the recurrent
connections of the hidden states, and bh is the bias vector for the hidden layer.

9
Fig. 8 The design presented by the authors adheres to a well-defined framework. Data on electricity
generation is acquired during Step 1. The second step entails preparing this data. During Step 3, the
features are retrieved and then transmitted through the BiLSTM for the purpose of decoding. In the
final step, Step 4, the model that has been trained is used to generate predictions. These predictions
are then assessed using simple error measures and shown visually through graphs. [30]
AB-Net demonstrated superior performance compared to other models such as BiL-
STM, CNN-BiLSTM, and an encoder-decoder (ED) in terms of mean absolute error
(MAE), mean squared error (MSE), and root mean square error (RMSE). The solar
dataset had mean squared error (MSE), root mean squared error (RMSE), and mean
absolute error (MAE) values of 0.0106, 0.1028, and 0.0743, respectively. In contrast,
the wind dataset had corresponding values of 0.0004, 0.0189, and 0.0109, respectively.
This hybrid network enhances energy management efficiency by facilitating enhanced
integration, trading, and control of renewable energy installations.

2.2 Big Data and Machine Learning for Smart Grid

Three layers compose the big data framework, as seen in Figure 9. The purpose of
the higher, simpler layer is to store, retrieve, and perform calculations on data. Data
management, sharing, and integration across many applications and domains are the

10
responsibilities of the intermediate layer; data privacy is a major concern in this layer.
The data mining platform uses data fusion technologies to preprocess data at the
lowest and deepest layer.

Fig. 9 A big data structure for a power utility using renewable energy [31].
The research work [31] explores the integration of big data analytics (BDA) in
smart grids and renewable energy power stations, focusing on predicting system sta-
bility. A framework is developed and applied to a dataset of 60,000 instances using
various machine learning methods, achieving high accuracy in predicting stability.
Three machine learning models -a decision tree, random forest classifier, and conven-
tional neural network (Deep learning) were utilized. The first model is related to the
classification model, where the aim was to identify if the smart grid system will be con-
sidered stable or not. A penalized linear regression has been employed by the fourth
machine learning model to estimate the stability differential equation root. Notably,
the penalized linear regression model attains 96% accuracy, while convolutional neu-
ral networks (CNN) achieve 87% accuracy for classification. The main contribution
lies in demonstrating the effectiveness of BDA in enhancing decision-making and sys-
tem stability in smart grid environments, paving the way for future research on larger
datasets and broader applications.

2.3 Using Machine Learning Models For Local Wind Power

Forecasting (Neural Network Architecture and Tree-Based
Approaches)
Local prosumers and consumers congregate in communities where they can collabo-
rate or engage in competition to achieve shared goals such as reducing greenhouse gas
emissions and/or power prices. Authors of the research, [32], propose data analytics
modules for Renewable Energy Communities (RECs) to optimize resource usage and
minimize electricity bills. It develops a day-ahead wind power forecasting algorithm
using Machine Learning techniques, improving accuracy by 10% and offering repre-
sentative consumption profiles for community members. The overall process is shown

11
in Figure 10, where abnormal wind power data is recognized in the target vector
and a ”forward pass” is performed to compute the actual forecast. The neural net-
work’s parameters are changed in the backward pass after the loss function has been
calculated in the third and final step.

Fig. 10 Method to train neural networks when anomalous data from wind power is present. [32].
The study contributes novel methods for wind power forecasting, abnormal data
handling, and generating consumption profiles, facilitating the integration of renewable
energy into community grids. Deployed in a pilot REC in Belgium, the modules aim to
align consumption with local generation, enhancing coordination and benefiting from
advantageous pricing mechanisms. On the E-Cloud pilot project data, the ENSEMBLE
model has demonstrated the greatest forecasting performance. Its output is calculated
as the average of the results of four different machine learning models: random forests,
gradient boosting decision trees, multilayer perceptrons, and bi-directional LSTMs.
The training procedure has been modified to account for the automated detection
of anomalous wind power data samples, which has significantly increased forecasting
accuracy. a method that uses dynamic time warping to create community members’
representative power usage profiles.

Fig. 11 Time series reveals the test set’s actual and predicted wind power for a single day. [32].

12
 In Figure 11, the actual wind power generation (in black) and the wind power
forecast (obtained with the ENSEMBLE model) as a function of time are illustrated
for a random day in the test set.

2.4 Forecasting Solar Radiation using Ensemble Feature

Selection and Machine Learning
Precise prediction of solar radiation is vital for the reliable operation of power systems
that rely heavily on photovoltaic generation. This study [33] assesses the performance
of machine learning (ML) methods in predicting solar radiation. The algorithms evalu-
ated include Support Vector Regression (SVR), Extreme Gradient Boosting (XGBT),
Categorical Boosting (CatBoost), and Voting-Average (VOA). A novel approach is
introduced to choose significant variables and their previous observations, hence
improving the accuracy of predicting. The ensemble feature selection approach com-
bines Pearson’s coefficient, mutual information, random forest, and relief to choose
the most related endogenous and exogenous inputs, together with their historical
observation values.The suggested method for estimating solar radiation consists of
five important phases, as illustrated in Figure 12. Tested on real data from Salvador,
Brazil, VOA outperforms other algorithms in all prediction time horizons. Contribu-
tions include comparing ML algorithms, introducing CatBoost for solar forecasting,
and proposing an ensemble feature selection method. Results show improved accuracy
and highlight the importance of selecting appropriate input variables and past obser-
vations for solar radiation forecasting. For the summer and winter datasets, VOA had
the faster training speed (183.93 s and 14.76 s, respectively). Being more sophisti-
cated, VOA combines SVR, XGBT, and CatBoost. Further research is recommended
to optimize parameter selection and explore applications in different locations and
forecasting domains.

Fig. 12 Diagram illustrating the suggested methods [33].

13
2.5 Optimal Hybrid AI Method (Time Delay Neural Network
(TDNN)) For Smart Grid Energy Management
In order to increase the effectiveness of energy management on a smart grid like Renew-
able Energy Community, authors of this research [34] provide an optimum hybrid AI
strategy in this work. In order to predict the future values of the energy characteris-
tics in the community, this technique uses a Time Delay Neural Network. Afterwards,
a stochastic model predictive control uses these predictions to optimize community
operations using an appropriate battery energy storage system management approach.
The Mean Absolute Error for solar generation, total energy consumption, and com-
mon services is 1.60 kW, 2.15 kW, and 0.30 kW, respectively, based on the outcomes
of forecasts made using a public dataset with a 24-hour prediction horizon. In compar-
ison to its rivals, the model predictive management fed with such forecasts yields the
highest revenue. When compared to using the exact same system of management with-
out taking use of forecasts from a forecasting approach, the total income is enhanced
by 18.72%.

2.6 Techniques for Detecting Transient Stability Using

CNN-LSTM
When a disturbance arises, electric power systems are more vulnerable to instability
because they function closer to their limitations. Problems with transient stability are
particularly common. [35], the research introduces a novel CNN-LSTM model (shown
in Figure 13) to detect transient stability in electric power systems using historical
data events. Addressing complexities in power system operation, the model considers
noise, delay, loss in measurement data, line outage, and variable renewable energy
(VRE) integration scenarios. Achieving over 99% accuracy in identifying stable and
unstable conditions, the proposed model outperforms conventional CNN and LSTM
approaches in terms of computation time, averaging 190.4 seconds. By training on
synthetic data and considering various system parameters, this model demonstrates
promise for enhancing transient stability detection in real-world power systems, albeit
with challenges in implementation and optimization. In terms of computation time
and performance, the CNN-LSTM approach outperforms the CNN and convLSTM
approaches.

14
Fig. 13 CNN-LSTM design that is being proposed [35].

3 More Application of AI in Renewable Energy

Domain
The energy industry’s significant development in the use of AI, IoT, and smart grid
infrastructure control has led to a study of cyber-attacks and a rapid rise in the num-
ber of solutions [36]. When it comes to load balancing renewable energy, automation is
particularly crucial since it ensures grid stability by matching distributed generation
to load [37]. The actions of rogue or rouge actors combined with inadequate sys-
tem design, implementation, or configuration pose a threat to load balancing and the
smart grid as a whole. Before a cybercatastrophy occurs, it is imperative to perform a
thorough investigation about the viability and urgency of these possible dangers [37].

3.1 Implementing AI to predict renewable power

Researchers and the energy business have paid close attention to big data as well as
artificial intelligence (AI) in recent years. This interest stems from two key factors:
(i) growing processing power and (ii) the exponential global data expansion [38]. It is
suggested to use Artificial Neural Networks (ANNs) for modeling and sizing a clean
energy community (CEC) that uses energy storage devices, electric car charging sta-
tions, and a PV-wind hybrid system to fulfill building district energy demand by the
authors of reserach [38]. While the second ANN is used to estimate the grid energy
indication factor, the first ANN is utilized to anticipate the energy performance indi-
cators, which include the fulfilled load fraction and the utilization factor of the energy
generated. ANNs are trained on a vast amount of data in all weather scenarios, with
adjustable power system configurations, and with a range of electrical demands. They
use advanced models of each CEC component to anticipate the year CEC energy per-
formance directly, without doing any system dynamic simulations. The impact of each

15
input on each output was assessed using the Garson algorithm. Compared to those
employed in ANN training, optimized ANNs have a very high prediction accuracy of
CECs since they are made up of a single hidden layer with twenty neurons.

3.2 Preventing Cyberattacks in Renewable Energy Field

The clever renewable energy to cyberattacks, opening a new avenue for hackers to
exploit the weaknesses in renewable energy [39]. This clearly impacts smart renewable
energy’s operational and economical aspects [39]. A number of contradicting Cyber-
Physical Power Systems (CPPSs) have seen moderate modifications in new kinds
of power systems due to advancements in communication networks, monitoring and
smart control systems, as well as the widespread usage of Internet-based structures
[40]. Cyber components and power portions are typically connected in these archi-
tectures. CPPSs deal with recently developed problems like as security, vulnerability,
resilience, stability, and dependability. Assessing, evaluating, and offering remedies to
lessen or resolve these issues heavily rely on precise modeling techniques and investigat-
ing the interplay processes related to Smart Grids’ (SGs’) cyber-security. Many studies
focused on creating reliable methods for identifying and/or defending against cyber-
attacks. These included studies on detection mechanisms that use federated learning
techniques, perfect or imperfect false data attacks, false data-driven or data-injection
attacks, and designing resilient communication networks [40].
The research article [41] introduces a novel data-driven approach for detecting false
data injection (FDI) attacks in energy forecasting systems, crucial for the reliabil-
ity of power systems amidst increasing reliance on renewable energy and information
technology. Utilizing cross-validation, least-squares, and z-score metrics, the proposed
mechanism effectively identifies intrusions without relying on system models or param-
eters, enhancing system resilience. Through experiments with tree-based wind power
forecasting models, the approach successfully detects and removes corrupted data,
restoring forecasting accuracy and generalizability.
The increasing availability of energy storage devices and the growing use of inter-
mittent renewable energy supplies are causing significant changes in power systems
[42]. These developments prompt smart-grid operators to imagine a day when micro-
grids would facilitate peer-to-peer energy trading, which will result in the creation
of transactive energy systems. Due in major part to their remarkable durability,
blockchains have attracted a lot of interest from academics and business for their
possible use in decentralized TES. The research [42] investigates blockchain-based
Transactive Energy Systems (TES) and introduces a novel class of attacks target-
ing gateways between prosumers and the system. The decentralized TES’s general
architecture is depicted in Figure 14.The study formulates a comprehensive model
of blockchain-based TES, analyzes various threat scenarios, and proposes attack mit-
igation strategies. Through experiments using GridLAB-D and a private Ethereum
network, the research demonstrates that even simple attacks can manipulate market
equilibria, but mitigation measures like detection and gateway switching effectively
counteract these threats, ensuring the integrity and resilience of the system.

16
Fig. 14 Infrastructure for the decentralized transactive energy system [42].

3.3 Optimizing consumption of energy employing artificial

intelligence
The subject matter ”Cost Savings” via AI-powered energy efficiency in smart systems
is one of the advantages. Lower operating expenses and electricity bills are directly
correlated with decreased energy use. A variety of methods are included in AI tech-
nology, such as predictive modeling, data analytics, and machine learning. AI-powered
systems may learn from data trends, make wise judgments, and optimize energy use
in real time thanks to their capabilities. When it comes to optimizing energy usage
in smart buildings, AI has several advantages. It makes it possible to gather and ana-
lyze enormous volumes of data from different sensors, meters, and Internet of things
devices. One of the main causes of buildings’ energy consumption is their HVAC
(heating, ventilation, and air conditioning) systems. By taking into account variables
like occupancy, the state of the weather, and the need for thermal comfort, AI-based
algorithms can improve HVAC operations.
[43] This study proposes a reliable method to optimize energy consumption in
buildings using EnergyPlus software and a robust artificial neural network (ANN)
model trained with data from a research center in Iran. Genetic algorithms are used to
optimize a building’s energy usage. Highly efficient features are evaluated in the build-
ing’s energy usage. By conducting sensitivity analysis, the most influential parameters
on energy consumption are identified, with the number of occupants and wall U-
value being the most significant. The proposed ANN model accurately predicts energy
consumption, enabling optimization through a Genetic Algorithm-based approach,
resulting in a substantial 35% reduction in energy usage. Overall, the study provides
valuable insights into optimizing energy consumption in buildings, offering a promising
methodology for similar applications.

4 Challenges of AI adoption in the smart energy

industry
The smart energy industry faces a variety of bottleneck obstacles to AI adoption,
including poor or nonexistent data, problems with AI network parameters, difficulties
with technical infrastructure, a shortage of skilled professionals, integration difficulties,

17
risks, and legal and compliance concerns. For building energy systems, failure detec-
tion and diagnosis provide additional difficult issues [44]. Various studies acknowledge
that inadequate information and data insecurity are two of the main problems that
energy systems face [45]. The performance and dependability of the energy system
are impacted by subpar sensors, controllers, and controlled devices for data estima-
tion and energy system operation. The energy market presents additional difficulties
for the power grid’s intricate correlations and strong coupling, as well as for the high
data dimensionality and enormous complexity of large-scale simulation grid data [46].
For grid operations, integrating renewable energy sources like solar and wind with AI
is equally challenging and complex [47].
The adoption of artificial intelligence (AI) in the smart energy industry faces sev-
eral challenges. These include a lack of key AI skills among decision-makers, hindering
the understanding of AI’s potential benefits [24]. Additionally, there is a shortage
of practical expertise in developing reliable AI applications with real-world benefits,
leading to reluctance among power companies to adopt new strategies. Outdated
power system infrastructure [24]further complicates modernization efforts, with util-
ity companies struggling to manage vast amounts of dispersed and disorganized data.
Economic pressures also impede the integration of innovative energy technologies due
to high costs and resource requirements. Furthermore, decentralization and diversi-
fication of energy supply, coupled with emerging AI technologies, present complex
challenges for energy production and distribution worldwide. Dependency on cellular
technologies limits AI’s potential [24], especially in developing economies. Moreover,
AI-based applications often act as black boxes for consumers, raising concerns about
transparency and security, particularly in the face of increasing cyber threats in the
digital energy infrastructure. Addressing these challenges is essential to unlock the full
potential of AI in revolutionizing the smart energy industry.

5 Current Challenges and Future Research

Directions
The integration of artificial intelligence (AI) in renewable energy research poses various
challenges and opens avenues for future exploration. Firstly, the complexity and cost
of AI algorithms used in optimization, design, and management of renewable energy
systems need to be addressed. Simplifying and making these models cost-effective is
crucial for widespread adoption and scalability.
Addressing real-time issues such as automatically controlling load demand for wind
and solar power farms under extreme weather conditions, enhancing the efficiency of
geothermal plants using IoT and AI, automating circuit breakers and relays under
fault conditions, and preventing/detecting the impact of lightning on transmission
lines remain critical challenges.
Regulators and policymakers must recognize AI’s systemic impact on reducing
greenhouse gas emissions and ensure flexible policies to facilitate the rapid deployment
of AI technologies in the energy sector. Incorporating ethical implications of AI in
corporate social responsibility standards is essential for responsible AI deployment.
Establishing participatory mechanisms to understand stakeholder expectations and

18
limitations in AI application cases is imperative. Building AI operating systems that
accurately identify energy efficiency and cost savings for consumers and businesses
will be invaluable.
Furthermore, enhancing transparency and accessibility of system data, protecting
AI-based models from security risks, and integrating AI with other technologies like
robotics, sensors, and IoT for optimization are key areas for future research. With
distributed energy resources and microgrids, artificial intelligence (AI) may be a key
component of the grid’s integration of renewable energy sources and promote oper-
ational independence. AI may be introduced to developing power markets with the
help of international financial institutions, enabling a sustainable energy transition.
Last but not least, realizing AI’s promise in energy systems calls for a multifaceted
strategy, assessing the viability of integrating AI, finding interesting AI methodologies,
and offering practical insights for improving performance while pleasing a range of
multidisciplinary audiences. In order to solve the difficulties posed by global climate
change and achieve sustainable energy transitions, it is imperative that comprehensive
frameworks and techniques for the efficient integration of AI in energy systems be
developed.
From revolutionizing medical diagnoses [48], [49], [50], [51], [52] to enhancing per-
sonalized learning in education, and optimizing transportation networks for efficiency,
artificial intelligence has been instrumental across various sectors [53], [54]. Now, with
recent advancements, the power sector is harnessing AI’s potential to improve grid
management, forecast energy demands, and enhance overall operational efficiency.
Researchers anticipate that major gains in energy efficiency and sustainability will
result from the integration of artificial intelligence (AI) with renewable energy sources.
Artificial intelligence (AI) has the potential to increase forecasting accuracy, stabi-
lize the grid, optimize renewable energy systems, and provide more intelligent energy
management options. Given AI’s capacity for massive data analysis and real-time
decision-making, we see a future in which renewable energy sources will predominate
in supplying the world’s energy needs while minimizing its negative environmental
effects. A more robust and sustainable energy environment is anticipated as a result
of the convergence of AI with renewable energy, opening the door to a cleaner and
greener future.

Table 2: An overview of this survey article that includes the

methodology and results of previous studies

Reference Year Methodology Model Data Source Research Outcome

[30] 2021 Used deep learning (DL) method Solar Dataset [55], AB-Net approach
AB-Net for forecasting of renewable Wind Dataset [56] provides MSE
energy generation 0.0004, RMSE
0.0189.
Continued on next page

19
 Table 2 – continued from previous page
Reference Year Methodology Model Data Source Research Outcome
[31] 2022 Used decision tree, random forest, Smart Grid data [57] Linear regression
and conventional neural network, model attains 96%
linear regression for smart grid man- accuracy.
agement
[32] 2021 Developed a day-ahead wind power Meteo Data (Wind Ensemble model
forecasting algorithm using Machine speed, pressure, provided improved
Learning algoriths (random forests, temperature) [58], [32] RMSE 2327 kW.
gradient boosting decision trees,
multilayer perceptrons, and bi-
directional LSTMs) to minimize
electricity bills
[33] 2023 Extreme Gradient Boosting Brazilian National The training speed
(XGBT), Support Vector Regres- Institute of Meteorol- of VOA was greater
sion (SVR), Categorical Boosting ogy website (INMET) than that of XGBT.
(CatBoost), and Voting-Average [59] Winter-VOA fore-
(VOA), which combines SVR, casting accuracy
XGBT, and CatBoost, are the was 0.2417 MAE,
machine learning techniques that whereas summer-
are employed for forecasting of solar VOA forecasting
radiation accuracy was
0.2877 MAE.
[34] 2022 Time Delay Neural Network for Data of PV generation Total income is
energy management plant [60] enhancedby 18.72%
[35] 2022 CNN-LSTM model to detect tran- synthetic dataset Discovery of tran-
sient stability in electric power sys- generated from the sient stability
tems DigSILENT Power- exceeded 99%.
Factory simulation,
available at: [61]
[38] 2022 Artificial Neural Networks (ANNs) The computation of The R2 values for
for electric vehicle (EV) charging the eight dimension- the SLF, UF, and
station and wind-photovoltaic bat- less input data requires GEIF forecasts are,
tery HRES performance assessment the input climate data in order, 0.952,
from the 48 locales 0.925, and 0.942.
and the load data from
the three building dis-
tricts under considera-
tion [38]
[42] 2021 explored TES based on blockchain Historical data on the Threats reduced
and presented a new category of prosumers’ previous by switching the
cyberattacks bids gateway and using
detection.
Continued on next page

20
 Table 2 – continued from previous page
Reference Year Methodology Model Data Source Research Outcome
[43] 2020 Genetic algorithms are used to opti- After being gathered, Garson’s method
mize a building’s energy usage. weather information demonstrates the
Highly efficient features are evalu- was loaded into the ANN model’s
ated in the building’s energy usage. Metronome program excellent accu-
(”National and Local racy of 99.9%
Weather Radar, Daily when it comes to
Forecast, Hurricane remodeling.
and information from
The Weather Chan-
nel and weather.com,”
n.d.)

6 Conclusion
An extensive review of the use of artificial intelligence (AI) approaches in renew-
able energy (RE) systems is given in our survey article. Nine AI-based techniques for
modern power systems—which address topics including renewable power production,
energy forecasting, and system optimization—are found by means of a comprehensive
literature study. The reviewed research demonstrate how several AI methods, like as
neural networks, decision trees, random forests, and deep learning, are successful in
tackling a range of RE management concerns.
The surveyed research demonstrates the potential of AI in enhancing controlla-
bility, data handling, cyberattack prevention, and smart grid implementation within
the energy industry. Furthermore, the superiority of AI models over conventional
approaches is highlighted, particularly in terms of accuracy, efficiency, and adaptability
to complex scenarios. Future directions for the integration of AI in renewable energy
systems are outlined, emphasizing the need for continued research and development in
this rapidly evolving field. Key areas for future exploration include the refinement of AI
algorithms, the integration of AI with other emerging technologies such as blockchain
and IoT, and the establishment of regulatory frameworks to support AI adoption in
the energy sector.
Overall, this survey paper contributes valuable insights into the state-of-the-art
AI applications in renewable energy and provides a roadmap for future research
endeavors aimed at leveraging AI for sustainable energy transition and environmental
conservation.

Declarations
Funding
A fraction of the financial support for this research is provided by DOEd Grant
P116Z220008 (1) and NSF Grant No. 2306109. The author(s) presents all viewpoints,
discoveries, and conclusions; the sponsor(s) may not necessarily endorse them.

21
Conflict of interest
The authors have no conflicts of interest to declare.

Ethics approval
This review paper does not include any original research that involves human beings,
animal subjects, or sensitive data.

Consent to participate
The papers and data included in this review were obtained from publically accessible
and previously published literature.

Consent for publication

All authors mentioned in this review study have examined and endorsed the final work
for submission to the Machine Learning journal. Every author has made substantial
contributions to the research, writing, and revision of the manuscript. All writers have
thoroughly reviewed and approved the content included in this review, and they have
given their explicit agreement for its publication.

Availability of data and materials

The data and materials utilized are available in the references of this study.

Code availability
The review is conducted by thoroughly analyzing publically accessible literature, and
any mentions of particular approaches or algorithms are properly credited to their orig-
inal sources. We have made diligent efforts to include explicit citations and references
to facilitate readers in locating the source works for additional scrutiny.

Authors’ contributions
Conceptualization: [Parag Biswas], [Angona Biswas]; Methodology: [Abdur Rashid],
[Angona Biswas]; Formal analysis and investigation: [Angona Biswas], [Abdullah Al
Nasim]; Writing - original draft preparation: [Angona Biswas][Abdullah Al Nasim],
[Abdur Rashid],[Parag Biswas]; Writing - review and editing: [Abdur Rashid]; Fund-
ing acquisition: [Kishor Datta Gupta]; Resources: [Abdur Rashid], [Parag Biswas];
Supervision: [Kishor Datta Gupta]

References
[1] He, X., Khan, S., Ozturk, I., Murshed, M.: The role of renewable energy invest-
ment in tackling climate change concerns: Environmental policies for achieving
sdg-13. Sustainable Development 31(3), 1888–1901 (2023)

22
 [2] Lorente, D.B., Mohammed, K.S., Cifuentes-Faura, J., Shahzad, U.: Dynamic
connectedness among climate change index, green financial assets and renew-
able energy markets: Novel evidence from sustainable development perspective.
Renewable Energy 204, 94–105 (2023)

[3] Anser, M.K., Khan, K.A., Umar, M., Awosusi, A.A., Shamansurova, Z.: Formu-
lating sustainable development policy for a developed nation: exploring the role
of renewable energy, natural gas efficiency and oil efficiency towards decarboniza-
tion. International Journal of Sustainable Development & World Ecology 31(3),
247–263 (2024)

[4] Al-Ismail, F.S., Alam, M.S., Shafiullah, M., Hossain, M.I., Rahman, S.M.: Impacts
of renewable energy generation on greenhouse gas emissions in saudi arabia: A
comprehensive review. Sustainability 15(6), 5069 (2023)

[5] Yi, S., Abbasi, K.R., Hussain, K., Albaker, A., Alvarado, R.: Environmental con-
cerns in the united states: Can renewable energy, fossil fuel energy, and natural
resources depletion help? Gondwana Research 117, 41–55 (2023)

[6] Laskaratou, V.: Energy dependence and national security. Master’s thesis,
Πανεπιστ ήµιo Πειραιώς (2023)

[7] Bamisile, O., Cai, D., Adun, H., Taiwo, M., Li, J., Hu, Y., Huang, Q.: Geothermal
energy prospect for decarbonization, ewf nexus and energy poverty mitigation in
east africa; the role of hydrogen production. Energy Strategy Reviews 49, 101157
(2023)

[8] Qiu, L., Wang, X., Wei, J.: Energy security and energy management: The role of
extreme natural events. Innovation and Green Development 2(2), 100051 (2023)

[9] Li, F., Zhang, J., Li, X.: Energy security dilemma and energy transition policy
in the context of climate change: A perspective from china. Energy Policy 181,
113624 (2023)

[10] Ang, T.-Z., Salem, M., Kamarol, M., Das, H.S., Nazari, M.A., Prabaharan, N.: A
comprehensive study of renewable energy sources: Classifications, challenges and
suggestions. Energy Strategy Reviews 43, 100939 (2022)

[11] Rizzi, F., Eck, N.J., Frey, M.: The production of scientific knowledge on renew-
able energies: Worldwide trends, dynamics and challenges and implications for
management. Renewable Energy 62, 657–671 (2014)

[12] Ludin, N.A., Mustafa, N.I., Hanafiah, M.M., Ibrahim, M.A., Teridi, M.A.M.,
Sepeai, S., Zaharim, A., Sopian, K.: Prospects of life cycle assessment of renewable
energy from solar photovoltaic technologies: A review. Renewable and Sustainable
Energy Reviews 96, 11–28 (2018)

23
[13] International Renewable Energy Agency (IRENA) and Renewable Energy Policy
Network for the 21st Century (REN21): Global Renewable Energy Capacity by
Source (2020-2023). Data obtained from reports published by the International
Renewable Energy Agency (IRENA) and the Renewable Energy Policy Network
for the 21st Century (REN21). (2024)

[14] Unknown: Renewable Energy Share in Electricity. News. Accessed on February

7, 2019 (2019)

[15] Bhuiyan, M.A., An, J., Mikhaylov, A., Moiseev, N., Danish, M.S.S.: Renew-
able energy deployment and covid-19 measures for sustainable development.
Sustainability 13(8), 4418 (2021)

[16] BP: Statistical Review of World Energy 2020. https://www.bp.com/

content/dam/bp/business-sites/en/global/corporate/pdfs/energy-economics/
statistical-review/bp-stats-review-2020-full-report.pdf Accessed 2021-04-15

[17] Petrović, R.M., Kocić, N., Stojanović, R.B.: The importance of renewable energy
sources for sustainable development. Economics of Sustainable Development 4(2),
15–24 (2020)

[18] Perea-Moreno, A.-J.: Renewable energy and energy saving: Worldwide research
trends. MDPI (2021)

[19] Alam, M.M., Murad, M.W.: The impacts of economic growth, trade openness
and technological progress on renewable energy use in organization for economic
co-operation and development countries. Renewable Energy 145, 382–390 (2020)

[20] Routledge, T..F.J.: Combating climate change: The role of science and engineering
(2021)

[21] Onwusinkwue, S., Osasona, F., Ahmad, I.A.I., Anyanwu, A.C., Dawodu, S.O.,
Obi, O.C., Hamdan, A.: Artificial intelligence (ai) in renewable energy: A review
of predictive maintenance and energy optimization (2024)

[22] Karim, T.F., Lipu, M.H., Mahmud, M.S.: Electricity access improvement using
renewable energy and energy efficiency: a case of urban poor area of dhaka,
bangladesh. International Journal of Renewable Energy Research (IJRER) 7(3),
1296–1306 (2017)

[23] Tao, Y., Destek, M.A., Pata, U.K., Khan, Z.: Environmental regulations and
carbon emissions: the role of renewable energy research and development expen-
ditures. Sustainability 15(18), 13345 (2023)

[24] Ahmad, T., Zhang, D., Huang, C., Zhang, H., Dai, N., Song, Y., Chen, H.:
Artificial intelligence in sustainable energy industry: Status quo, challenges and
opportunities. Journal of Cleaner Production 289, 125834 (2021)

24
[25] Georgiou, M.: The role of ai technology in improving the renewable energy sector.
Insider (2019)

[26] Lahmadi, K.: Artificial intelligence can improve renewable energy production,
report shows. Morocco World News (2023)

[27] Khan, N., Haq, I.U., Khan, S.U., Rho, S., Lee, M.Y., Baik, S.W.: Db-net: A novel
dilated cnn based multi-step forecasting model for power consumption in inte-
grated local energy systems. International Journal of Electrical Power & Energy
Systems 133, 107023 (2021)

[28] Nam, K., Hwangbo, S., Yoo, C.: A deep learning-based forecasting model for
renewable energy scenarios to guide sustainable energy policy: A case study of
korea. Renewable and Sustainable Energy Reviews 122, 109725 (2020)

[29] Gensler, A., Henze, J., Sick, B., Raabe, N.: Deep learning for solar power fore-
casting—an approach using autoencoder and lstm neural networks. In: 2016
IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp.
002858–002865 (2016). IEEE

[30] Khan, N., Ullah, F.U.M., Haq, I.U., Khan, S.U., Lee, M.Y., Baik, S.W.: Ab-
net: A novel deep learning assisted framework for renewable energy generation
forecasting. Mathematics 9(19), 2456 (2021)

[31] Mostafa, N., Ramadan, H.S.M., Elfarouk, O.: Renewable energy management in
smart grids by using big data analytics and machine learning. Machine Learning
with Applications 9, 100363 (2022)

[32] Grève, Z.D., Bottieau, J., Vangulick, D., Wautier, A., Dapoz, P.-D., Arrigo,
A., Toubeau, J.-F., Vallée, F.: Machine learning techniques for improving
self-consumption in renewable energy communities. Energies 13(18), 4892 (2020)

[33] Solano, E.S., Dehghanian, P., Affonso, C.M.: Solar radiation forecasting using
machine learning and ensemble feature selection. Energies 15(19), 7049 (2022)

[34] Conte, F., D’Antoni, F., Natrella, G., Merone, M.: A new hybrid ai optimal man-
agement method for renewable energy communities. Energy and AI 10, 100197
(2022)

[35] Azhar, I.F., Putranto, L.M., Irnawan, R.: Development of pmu-based tran-
sient stability detection methods using cnn-lstm considering time series data
measurement. Energies 15(21), 8241 (2022)

[36] Sani, A.S., Yuan, D., Jin, J., Gao, L., Yu, S., Dong, Z.Y.: Cyber security frame-
work for internet of things-based energy internet. Future Generation Computer
Systems 93, 849–859 (2019)

25
[37] Vernotte, A., Välja, M., Korman, M., Björkman, G., Ekstedt, M., Lagerström, R.:
Load balancing of renewable energy: a cyber security analysis. Energy Informatics
1, 1–41 (2018)

[38] Mazzeo, D., Herdem, M.S., Matera, N., Bonini, M., Wen, J.Z., Nathwani, J.,
Oliveti, G.: Artificial intelligence application for the performance prediction of a
clean energy community. Energy 232, 120999 (2021)

[39] Rekeraho, A., Cotfas, D.T., Cotfas, P.A., Bălan, T.C., Tuyishime, E., Acheam-
pong, R.: Cybersecurity challenges in iot-based smart renewable energy. Interna-
tional Journal of Information Security 23(1), 101–117 (2024)

[40] Ghiasi, M., Niknam, T., Wang, Z., Mehrandezh, M., Dehghani, M., Ghadimi, N.:
A comprehensive review of cyber-attacks and defense mechanisms for improving
security in smart grid energy systems: Past, present and future. Electric Power
Systems Research 215, 108975 (2023)

[41] Ahmadi, A., Nabipour, M., Taheri, S., Mohammadi-Ivatloo, B., Vahidinasab, V.:
A new false data injection attack detection model for cyberattack resilient energy
forecasting. IEEE Transactions on Industrial Informatics 19(1), 371–381 (2022)

[42] Barreto, C., Eghtesad, T., Eisele, S., Laszka, A., Dubey, A., Koutsoukos, X.:
Cyber-attacks and mitigation in blockchain based transactive energy systems. In:
2020 IEEE Conference on Industrial Cyberphysical Systems (ICPS), vol. 1, pp.
129–136 (2020). IEEE

[43] Ilbeigi, M., Ghomeishi, M., Dehghanbanadaki, A.: Prediction and optimization
of energy consumption in an office building using artificial neural network and a
genetic algorithm. Sustainable Cities and Society 61, 102325 (2020)

[44] Guo, Y., Wang, J., Chen, H., Li, G., Huang, R., Yuan, Y., Ahmad, T., Sun,
S.: An expert rule-based fault diagnosis strategy for variable refrigerant flow air
conditioning systems. Applied Thermal Engineering 149, 1223–1235 (2019) https:
//doi.org/10.1016/j.applthermaleng.2018.12.132

[45] Zhao, Y., Xiao, F., Wang, S.: An intelligent chiller fault detection and diagnosis
methodology using bayesian belief network. Energy and Buildings 57, 278–288
(2013) https://doi.org/10.1016/j.enbuild.2012.11.007

[46] Tang, Y., Huang, Y., Wang, H., Wang, C., Guo, Q., Yao, W.: Framework for
artificial intelligence analysis in large-scale power grids based on digital simu-
lation. CSEE Journal of Power and Energy Systems 4, 459–468 (2018) https:
//doi.org/10.17775/cseejpes.2018.01010

[47] Puri, V., Jha, S., Kumar, R., Priyadarshini, I., Hoang Son, L., Abdel-Basset,
M., Elhoseny, M., Viet Long, H.: A hybrid artificial intelligence and internet
of things model for generation of renewable resource of energy. IEEE Access 7,

26
 111181–111191 (2019) https://doi.org/10.1109/access.2019.2934228

[48] Al Nasim, M.A., Al Munem, A., Islam, M., Palash, M.A.H., Haque, M.M.A.,
Shah, F.M.: Brain tumor segmentation using enhanced u-net model with empir-
ical analysis. In: 2022 25th International Conference on Computer and Infor-
mation Technology (ICCIT), pp. 1027–1032 (2022). https://doi.org/10.1109/
ICCIT57492.2022.10054934

[49] Biswas, A., Islam, M.S.: A hybrid deep cnn-svm approach for brain tumor clas-
sification. Journal of Information Systems Engineering & Business Intelligence
9(1) (2023)

[50] Biswas, A., Islam, M.S.: Mri brain tumor classification technique using fuzzy c-
means clustering and artificial neural network. In: Ibrahim, R., Porkumaran, K.,
Kannan, R., Mohd Nor, N., Prabakar, S. (eds.) International Conference on Arti-
ficial Intelligence for Smart Community. Lecture Notes in Electrical Engineering,
vol. 758. Springer, ??? (2022). https://doi.org/10.1007/978-981-16-2183-3 95 .
https://doi.org/10.1007/978-981-16-2183-3 95

[51] Biswas, A., Abdullah Al, N.M., Ali, M.S., Hossain, I., Ullah, M.A., Talukder,
S.: Active learning on medical image. In: Data Driven Approaches on Medical
Imaging, pp. 51–67. Springer, ??? (2023)

[52] Biswas, A., Md Abdullah Al, N., Imran, A., Sejuty, A.T., Fairooz, F., Puppala,
S., Talukder, S.: Generative adversarial networks for data augmentation. In: Data
Driven Approaches on Medical Imaging, pp. 159–177. Springer, ??? (2023)

[53] Irbaz, M.S., Al Nasim, M.A., Ferdous, R.E.: Real-time face recognition system
for remote employee tracking. In: Arefin, M.S., Kaiser, M.S., Bandyopadhyay,
A., Ahad, M.A.R., Ray, K. (eds.) Proceedings of the International Conference on
Big Data, IoT, and Machine Learning. Lecture Notes on Data Engineering and
Communications Technologies, vol. 95. Springer, ??? (2022). https://doi.org/10.
1007/978-981-16-6636-0 13 . https://doi.org/10.1007/978-981-16-6636-0 13

[54] Biswas, A., Islam, M.S.: An efficient cnn model for automated digital handwrit-
ten digit classification. Journal of Information Systems Engineering and Business
Intelligence 7(1), 42–55 (2021)

[55] DATA.GO.KR. Accessed on 5 April 2021

[56] NREL Wind Prospector. Accessed on 5 April 2021

[57] Arzamasov, V., Böhm, K., Jochem, P.: Towards concise models of grid stabil-
ity. In: 2018 IEEE International Conference on Communications, Control, and
Computing Technologies for Smart Grids (SmartGridComm), pp. 1–6 (2018).
https://doi.org/10.1109/SmartGridComm.2018.8587498

27
[58] Shen, X., Fu, X., Zhou, C.: A combined algorithm for cleaning abnormal data of
wind turbine power curve based on change point grouping algorithm and quartile
algorithm. IEEE Transactions on Sustainable Energy 10(1), 46–54 (2018)

[59] INMET. Instituto Nacional de Meteorologia. Accessed on 23 October 2021. https:

//portal.inmet.gov.br/ Accessed 2021-10-23

[60] Ramos, S., Soares, J., Foroozandeh, Z., Tavares, I., Vale, Z.: Energy consumption
and PV generation data of 15 prosumers (15 minute resolution). Zenodo. Accessed
on [date] (2021). http://dx.doi.org/10.5281/zenodo.5106455

[61] Izzuddinfa: Transient Stability. Kaggle. https://www.kaggle.com/datasets/

izzuddinfa/transient-stability (year)

28